Performance Analysis of OFDM Systems in The Presence of Doppler-Effect
Performance Analysis of OFDM Systems in The Presence of Doppler-Effect
IOSR Journal of Electrical and Electronics Engineering (IOSR-JEEE)
e-ISSN: 2278-1676 Volume 4, Issue 3 (Jan. - Feb. 2013), PP 24-27
www.iosrjournals.org
www.iosrjournals.org 24 | Page
Performance analysis of OFDM systems in the presence of
Doppler-effect
Mohamed Tayebi
1
, Merahi Bouziani
1
1
(Telecommunications and Digital Signal Processing Laboratory, Djillali LiabesUniversity, Algeria)
Abstract : Carrier Frequency Offset (CFO) in Orthogonal Frequency Division Multiplexing (OFDM) systems
is generated partly by imperfections of the local oscillators (static CFO) and secondly, by the Doppler-effect
(dynamic CFO). The aim of the present study is to investigate the impact of Doppler-effect on OFDM
performances. A new expression of the global CFO is derived and Bit Error Rate performance is obtained by
Monte-Carlo simulation.
Keywords - Orthogonal Frequency Division Multiplexing (OFDM), Carrier Frequency Offset (CFO), Doppler-
effect, Inter Carriers Interferences (ICI), Carrier to Interferences Ratio (CIR).
I. Introduction
OFDM is simply implemented using Fast Fourier Transform (FFT) and presents innumerable
advantages such as high spectral efficiency and robustness against multipath fading. These advantages have
enabled telecommunication systems to make a significant forward step and access gain to new perspectives.
OFDM has found its place in optical communications [1] and in a large variety of broadcast standards such as
Digital Audio Broadcasting (DAB), Digital Video Broadcasting (DVB). However, its main disadvantage is its
sensitivity to Carrier Frequency Offset (CFO). This is mainly due to imperfections in local oscillators and
Doppler-effect present in mobile radio channels [2]. CFO gives rise to inter-carrier interference responsible for
the degradation of system performance[3,4,5]. For a transmitter-receiver pair, the offset due to the imperfections
of the local oscillators is constant, while the shift due to the Doppler- effect is variable and random. We can talk
about static CFO and dynamic CFO. This letter concerns the dynamic CFO where a new expression of the
global CFO is derived and in order to evaluate the OFDM performances, Bit Error Rates are obtained.
II. Doppler-effect and Carrier Frequency Offset analysis
The OFDM symbol is transmitted by the radio frequency circuit. The carrier is noted f
p
. Assume that
local oscillators of the transmitter and of the receiver do not oscillate at the same frequency, the frequency at the
receiver is:
diff p r
f f f + = (1)
f
diff
is the frequency difference between the transmitter and the receiver.
This frequency difference caused by the
imperfections of the local oscillators is constant and still exists. The normalized value is equal to:
diff diff
Tf = c (2)
When a mobile is moving at a speed v, a frequency shift f
d
occurs. This shift is called the Doppler-effect. Its
mathematical expression is given by:
o cos
c
v
f f
p d
= (3)
where c is the light speed, f
p
is the carrier frequency and o is the angle between the velocity vector and the
direction of the electromagnetic wave. The normalized value is:
o c cos
c
v
Tf
p d
= (4)
Figure 1 shows that the normalized Doppler frequency increases with velocity v and the angle o determines its
sign. For a mobile receiver, the received frequency is given by:
( )
|
.
|
\
|
+ + = o cos 1
c
v
f f f
diff p r
(5)
Performance analysis of OFDM systems in the presence of Doppler-effect
www.iosrjournals.org 25 | Page
The total carrier frequency offset will be:
|
.
|
\
|
+ + = o o cos 1 cos
c
v
f
c
v
f f
diff p T
(6)
Its normalized value is equal to:
diff p diff p T
c
v
Tf
c
v
c
v
Tf c o o c o c + ~
|
.
|
\
|
+ + = cos cos 1 cos (7)
-200
0
200
0
5
10
-0.5
0
0.5
o ()
v (m/s)
c
d
Fig.1: Normalized Doppler frequency
III. Inter-Carrier Interference and performance analysis
3.1 Inter-Carrier Interference analysis
The orthogonality between the sub-carriers of the OFDM signal is broken, which leads to the rise of
inter-carrier interference. The amplitude of ICI coefficients between the l
th
and the k
th
subcarriers is given by [4]:
|
.
|
\
|
+ +
|
.
|
\
|
+ +
=
diff p
diff p
c
v
Tf k l
N
N
c
v
Tf k l
k l S
c o
t
c o t
cos sin
cos sin
) ( (8)
These coefficients depend on the relative speed of the mobile, the angle o and the subcarrier indexes.
3.2 Carrier to Interference Ratio
This section examines the performances in the presence of the Doppler effect in terms of the CIR
which is given by [4]:
2
1
1
2
) (
) 0 (
=
=
N
l
l S
S
CIR (9)
knowing that [5]:
1 ) (
2
1
0
=
=
N
l
l S (10)
and:
|
.
|
\
|
+
|
.
|
\
|
+
=
diff p
diff p
c
v
Tf
N
N
c
v
Tf
S
c o
t
c o t
cos sin
cos sin
) 0 (
(11)
Performance analysis of OFDM systems in the presence of Doppler-effect
www.iosrjournals.org 26 | Page
We can rewrite Eq(9) in the following form:
2
2
) 0 ( 1
) 0 (
S
S
CIR
= (12)
Figure 2 shows the CIR as a function of speed v and the angle o for T = 224.10
-6
s and f
p
= 5 Ghz. The CIR
decreases when the relative velocity v increases and when the angle o is in the range: | | 0 90 = o and
| | 180 90 = o . The CIR is maximum for o = 90 , meaning that the Doppler-effect is cancelled.
0
50
100
150
0
5
10
0
10
20
30
40
50
o ()
v (m/s)
C
I
R
(
d
B
)
Fig.2: Carrier to Interferences Ratio
3.3 Bit Error Rate (BER)
Using the Monte Carlo method, Figure 3 shows the evolution of the BER for different values of
Signal to Noise Ratio (SNR) given an angle of arrival o . The simulation was performed with the following
values (relative velocity v = 100m/s and the normalized frequency offset 2 . 0 = c ). For 0 = o , the Doppler
effect is completely absent. For 90 < o , the shift due to the Doppler effect is positive, whereas for 90 > o , it
is negative. In this case, the performance is improved since the global offset is reduced.
0 5 10 15 20 25 30
10
-5
10
-4
10
-3
10
-2
10
-1
10
0
SNRo
B
E
R
o = 90
o = 60
o = 120
Fig.3: BER versus Signal to Noise Ratio for different values of o .
IV. Conclusion
In this letter, a new expression for the global CFO is derived for OFDM systems. We deduce that the
system performance, in terms of BER, is degraded by the CFO due to imperfections of the local oscillators and
the CFO due to the Doppler effect contributes to an additional degradation if it is in the same direction as the
first. However, performance will be improved if the two CFOs are in opposite directions.
Performance analysis of OFDM systems in the presence of Doppler-effect
www.iosrjournals.org 27 | Page
REFERENCES
Journal Papers:
[1] K.Kikuchi, Digital coherent optical communication systems: fundamental and future prospects, IEIEC Electronics Express, 8(20),
2011, 1642-1662.
[2] Y. Zhao and S.-G. Hggman, Sensitivity to Doppler shift and carrier frequency errors in OFDM systemsThe consequences and
solutions, Proc. 46th IEEE Vehicular Technology Conference, Atlanta, GA, 1996, 15641568.
[3] R. Uma Mahesh and A. K. Chaturvedi, " Closed Form BER Expressions for BPSK OFDM Systems with Frequency Offset" IEEE
Communications Letters, 14(8), 2010, 731-733.
[4] Y. Zhao and S.-G.Haggman, Intercarrier interference self-cancellation scheme for OFDM mobile communication systems, IEEE
Transactions on Communications, 49(7), 2001, 11851191.
[5] J. Lee, H. Lou, D. Toumpakaris, and J. Cioffi, SNR Analysis of OFDM Systems in the Presence of Carrier Frequency Offset for
Fading Channels IEEE Transactions on wireless communications, 5(12), 2006, 33603364.
